Wem vertrauen wir mehr: Mensch oder KI? Nicht nur Schülerinnen und Schüler können Opfer des “Automation Bias” werden – auch Lehrkräfte sollten die Ergebnisse von KI-Tools reflektieren.
A short educational resource explaining automation bias, the tendency to trust suggestions and decisions made by machines more than one’s own judgement. It uses the example of a student trusting an AI chatbot despite contradictory research and highlights the importance of critically checking AI-generated information.
Why it matters for ENACT-AI
The resource directly addresses a key AI-literacy issue: over-reliance on AI outputs. It encourages students and teachers to question AI-generated information, verify results, and maintain independent judgement rather than delegating their thinking to AI systems.
Learning outcomes
Learners understand the concept of automation bias, recognize situations in which people may over-trust AI systems, understand that AI tools can provide incorrect or misleading information, develop habits of verifying AI-generated outputs, and reflect on the limits of delegating judgement to AI.
Suggested activity
Ask learners to pose a question to an AI chatbot and independently research the same question using reliable sources. Compare the AI answer with the research, identify inaccuracies or missing context, and discuss why someone might be inclined to trust the AI answer. Finish by creating a short checklist for critically evaluating AI-generated information.